R09 brickwalling

[it-goes-to-eleven]

I ran onstage for some afrobeat saturday and saturated/brickwalled the r09:

mk21 > v3 > microtrack@24/48 & r09 @ 24/48

The bass wasn't THAT heavy.  I had hoped to post a comp because I expected this show would demonstrate some gripes I've had with the r09's bottom end despite the v3>r09 sometimes being preferred over the v3's a/d.  I'll probably still post the comp to give an idea of how it sounds but I'll run it less hot next time. I can hear some distortion in the bass, especially later in the show when I increased the gain a few db. It isn't too bad.  The 722 was running off the mics in the middle of the floor so the bacon was brunged home..

During the show the r09 peak hold was showing -6 and I kept inching up the gain from 20 to 25 on the v3. I did notice that the gain didn't seem to be increasing 1db for each db added but I didn't see the significance.  Didn't ever clip the r09 or v3 but came within a quarter or half db on the mt source.  The highest peaks for the r09 were -3.68L and -3.44R.

I was running the r09 line in, trim 1, no agc, low gain.

A track from later in the show had a final RMS of -16 and the r09 -11. So about 5dB of compression.

   RMS
  level        peak       
-16.3335dBFS -0.1684dBFS microtrack.wav
-10.9785dBFS -0.2463dBFS r09.wav

[spyder9]

I ran onstage for some afrobeat saturday and saturated/brickwalled the r09:

mk21 > v3 > microtrack@24/48 & r09 @ 24/48

The bass wasn't THAT heavy.  I had hoped to post a comp because I expected this show would demonstrate some gripes I've had with the r09's bottom end despite the v3>r09 sometimes being preferred over the v3's a/d.  I'll probably still post the comp to give an idea of how it sounds but I'll run it less hot next time. I can hear some distortion in the bass, especially later in the show when I increased the gain a few db. It isn't too bad.  The 722 was running off the mics in the middle of the floor so the bacon was brunged home..

During the show the r09 peak hold was showing -6 and I kept inching up the gain from 20 to 25 on the v3. I did notice that the gain didn't seem to be increasing 1db for each db added but I didn't see the significance.  Didn't ever clip the r09 or v3 but came within a quarter or half db on the mt source.  The highest peaks for the r09 were -3.68L and -3.44R.

I was running the r09 line in, trim 1, no agc, low gain.

A track from later in the show had a final RMS of -16 and the r09 -11. So about 5dB of compression.

   RMS
  level        peak       
-16.3335dBFS -0.1684dBFS microtrack.wav
-10.9785dBFS -0.2463dBFS r09.wav

Wow Brian!!  I've never brickwalled the R09 while using the Aerco.  And I've run hot, right up to zero, many times and never had a problem.  Hmmm...  sure its not the V3 analog?

I'm running an R-1 while the R09 gets fixed.  The R09 has a superior ADC to the R-1.  R-1 shows too much color and harshness in my recordings.  If you go on Archive and search spyder9, listen to my shows posted after 4/08/07 and you''ll get an idea.  Its like night and day. 

[rePat]

I was running the r09 line in, trim 1, no agc, low gain.

When you say Trim 1, is that the level setting on the R-09?  I read somewhere that "Unity" on Line In was between 7 and 9. 

[guysonic]

I was running the r09 line in, trim 1, no agc, low gain.

When you say Trim 1, is that the level setting on the R-09?  I read somewhere that "Unity" on Line In was between 7 and 9. 

It's true that BELOW some trim number the deck is attenuating the input signal.  Looking at the input sensitivity chart on my review page, and figuring the deck cannot handle more that 1 volt rms signal internally, there's at least about 12 dB of signal reduction at work at trim setting #1

Review signal input chart at: www.sonicstudios.com/r-09revw.htm#inputs

[it-goes-to-eleven]

It's true that BELOW some trim number the deck is attenuating the input signal.  Looking at the input sensitivity chart on my review page, and figuring the deck cannot handle more that 1 volt rms signal internally, there's at least about 12 dB of signal reduction at work at trim setting #1

You are correct... I had thought that 1 might be closer to true unity but it isn't and I was wrong about that. With some bench testing I just confirmed things get weird down there.

Squeezebox3 > v3 > r09

At trim 8 things behaved pretty much as expected when the v3 gain was increased. Peak levels would go up approx 5dB with each 5dB of v3 gain.

At 25 dB of gain and trim 8, the peak light was very infrequently lighting.
At trim 1, it took 45 dB of gain on the v3 to get some clip light action.

So there does appear to be some weird compression going on at trim 1. That said, it will be interesting to see if the r09 still beats out the v3 a/d on subject sound quality in the comp

[it-goes-to-eleven]

I did it to myself again last night....  I was running mk4 > psp2 > r09...  Since the r09 input is unbalanced, I hoped that would provide sufficient attenuation of the minimum 20dB of gain from the psp2.. Nope.. Loud show.. clipping..

Torn over whether to let it ride and just tape over the clip light so it wouldn't bug me, or drop the trim below 8...  Ultimately I "had to" drop the trim down to 1-2...  I'm sure hpf on the psp2 would have helped but I didn't want to do that. If I'd brought the necessary cable, I would have run the aerco into the r09 (since the gain goes down to 10 or 0) but that was not in my pre-show plan so I didn't bring it..

I had another source so I wasn't terribly concerned (otherwise I probably would have gone for some hpf). I also might have moved the MGs to the psp2/r09 with the -10db mic pad but the mics were already on the ceiling with the pad off....

[spyder9]

Flip the R09 over see what gain setting you have.  The gain setting is on the back of the R09.  If it says "high", drop that puppy down to "low"

[guysonic]

Flip the R09 over see what gain setting you have.  The gain setting is on the back of the R09.  If it says "high", drop that puppy down to "low"

If using the LINE input, High/Low switch should have NO effect as it works only with MIC input.

See input chart for settings effect at: www.sonicstudios.com/r-09revw.htm#inputs

[it-goes-to-eleven]

Definitely low gain... checked it before and even during the show...  They are pretty easy to flip in handling so I always check them and have been known to tape them. Guy says it doesn't matter but paranoia/caution rules on these things..

But this has gotten weirder.  This recording was horribly blown out. I figured the clips would just be clips and I'd salvage the bits that weren't clipped.. But chunks were dropped out.  It was so bad I suspected an r09 line jack failure but testing confirmed that isn't the problem. There is something going on with the psp2 to r09 interaction.

I did a bunch of testing and apparently the psp2 does not like to drive an unbalanced cable on the balanced outputs at high levels. It does just fine with balanced output.  I was able to reproduce the problem using the microtrack. I tried putting an inline unbal attenuator in front of the r09 and microtrack but saw the same problem.

So in order to get an idea how hot the levels must be to cut out the psp2, I substituted the 722 (balanced) for the r09.  With the 722 at -6, I see a level of about -4 on the 722.  Since the 722 can take a +26dBU signal (at -6) before clipping, I'm thinking the trouble point occurs at around +20dBu of psp2 output. I was never able to reproduce the problem with the psp2 driving balanced loads (722).

Of course that's past where the r09 would be clipping but the cut-out gotcha is much worse than transient clips. And it is a bummer that an attenuator cable on the r09 side can't fix it. Of course I'm also concerned that the psp2 doesn't "like" this config and that I may damage it.

I wonder if a different unbal cable config could solve this?

The cable is 24" long and was made with 1804a quad, wired like fig 2:

http://www.vandenhul.com/artpap/wiring4.htm

[Church-Audio]

Definitely low gain... checked it before and even during the show...  They are pretty easy to flip in handling so I always check them and have been known to tape them. Guy says it doesn't matter but paranoia/caution rules on these things..

But this has gotten weirder.  This recording was horribly blown out. I figured the clips would just be clips and I'd salvage the bits that weren't clipped.. But chunks were dropped out.  It was so bad I suspected an r09 line jack failure but testing confirmed that isn't the problem. There is something going on with the psp2 to r09 interaction.

I did a bunch of testing and apparently the psp2 does not like to drive an unbalanced cable on the balanced outputs at high levels. It does just fine with balanced output.  I was able to reproduce the problem using the microtrack. I tried putting an inline unbal attenuator in front of the r09 and microtrack but saw the same problem.

So in order to get an idea how hot the levels must be to cut out the psp2, I substituted the 722 (balanced) for the r09.  With the 722 at -6, I see a level of about -4 on the 722.  Since the 722 can take a +26dBU signal (at -6) before clipping, I'm thinking the trouble point occurs at around +20dBu of psp2 output. I was never able to reproduce the problem with the psp2 driving balanced loads (722).

Of course that's past where the r09 would be clipping but the cut-out gotcha is much worse than transient clips. And it is a bummer that an attenuator cable on the r09 side can't fix it. Of course I'm also concerned that the psp2 doesn't "like" this config and that I may damage it.

I wonder if a different unbal cable config could solve this?

The cable is 24" long and was made with 1804a quad, wired like fig 2:

http://www.vandenhul.com/artpap/wiring4.htm

Forgive my ignorance but what is a psp2 and what type of balanced output does it have? Transformer? electronic? If its electronic you should tie pin 3&1 together providing that the output of the preamp is pin 2 hot... If its not then pin 2&1 should be tied together if its a transformer output you can use pin 2 as hot pin 3 as negative and pin 1 floating connected to one side of the shield at the output end of the chain. I suspect it might be that this preamp is pin 3 hot not pin 2 if that is the case you would increase the distortion of the preamp significantly by tying pin 3 to ground. Not to mention the fact that you would be inverting the output 180 degrees.

Chris

[udovdh]

I was running the r09 line in, trim 1, no agc, low gain.

Why do that? (trim 1)
(apart from teh level 8 is unity hype)

You know that level 1 is maximum attenuation for the R09. So why not lower the level on a regular basis?
Maybe I am missing a point so please explin if that is the case.

[it-goes-to-eleven]

Forgive my ignorance but what is a psp2 and what type of balanced output does it have? Transformer? electronic? If its electronic you should tie pin 3&1 together providing that the output of the preamp is pin 2 hot... If its not then pin 2&1 should be tied together if its a transformer output you can use pin 2 as hot pin 3 as negative and pin 1 floating connected to one side of the shield at the output end of the chain. I suspect it might be that this preamp is pin 3 hot not pin 2 if that is the case you would increase the distortion of the preamp significantly by tying pin 3 to ground. Not to mention the fact that you would be inverting the output 180 degrees.

Thanks Chris!

The psp2 is a preamp that was made by a now defunct French company so there really isn't any factory support. The inputs are transformer based but I don't *think* the outputs are.

I have not heard distortion in the samples that were at a reasonable level. They sound good.

My cable currently ties 1&3 together.  Any suggestion on how I can determine which of 2 or 3 are hot using a DVM? I thought they were both hot in a balanced config with one of them being out of phase. I just tried to measure the dc voltage between 1&2 and 1&3 (nothing on the inputs). They both measured about .13 volts and were the same polarity (positive with ground neg).

Now... I did find this troubling comment about cabling the psp2 unbalanced.. it definitely gives me pause and I am trying to interpret it. I had read it before but my initial results with my unbal cable were good so I dismissed it:

one thing about the xlr interconnects for the psp2. This was relayed to me by Art Munson or Frank , I can't remember which. I've had two of these things and I fried the first one. I traded the hulk in on a NOS unit that was found for me. I got a call at work one day and Art/Frank told me to make sure that only two wires were soldered to the xlr connector. The third one was left free. I'm no xlr expert so details are fuzzy.

When they say two wires only, I assume they mean 1 to shield and 2 as the signal with 3 being left open (could be 2 open, 3 sig)..  So that implies the shield is connected at both ends.  Wondering if shield connected at both ends is preferable to shield tied at one end only and using a wire for the ground.

I use the 1804a for this because it is small and nice to use but I suppose an unbalanced cable would be better?  If I'm only using one wire and shield I imagine the quad is not ideal.

The psp2 does have an 1/8 unbalanced output and, gee, it does seem to work. I assumed I would get higher quality output with my own cable on the separate bal outs.

[Church-Audio]

Forgive my ignorance but what is a psp2 and what type of balanced output does it have? Transformer? electronic? If its electronic you should tie pin 3&1 together providing that the output of the preamp is pin 2 hot... If its not then pin 2&1 should be tied together if its a transformer output you can use pin 2 as hot pin 3 as negative and pin 1 floating connected to one side of the shield at the output end of the chain. I suspect it might be that this preamp is pin 3 hot not pin 2 if that is the case you would increase the distortion of the preamp significantly by tying pin 3 to ground. Not to mention the fact that you would be inverting the output 180 degrees.

Thanks Chris!

The psp2 is a preamp that was made by a now defunct French company so there really isn't any factory support. The inputs are transformer based but I don't *think* the outputs are.

I have not heard distortion in the samples that were at a reasonable level. They sound good.

My cable currently ties 1&3 together.  Any suggestion on how I can determine which of 2 or 3 are hot using a DVM? I thought they were both hot in a balanced config with one of them being out of phase. I just tried to measure the dc voltage between 1&2 and 1&3 (nothing on the inputs). They both measured about .13 volts and were the same polarity (positive with ground neg).

Now... I did find this troubling comment about cabling the psp2 unbalanced.. it definitely gives me pause and I am trying to interpret it. I had read it before but my initial results with my unbal cable were good so I dismissed it:

one thing about the xlr interconnects for the psp2. This was relayed to me by Art Munson or Frank , I can't remember which. I've had two of these things and I fried the first one. I traded the hulk in on a NOS unit that was found for me. I got a call at work one day and Art/Frank told me to make sure that only two wires were soldered to the xlr connector. The third one was left free. I'm no xlr expert so details are fuzzy.

When they say two wires only, I assume they mean 1 to shield and 2 as the signal with 3 being left open (could be 2 open, 3 sig)..  So that implies the shield is connected at both ends.  Wondering if shield connected at both ends is preferable to shield tied at one end only and using a wire for the ground.

I use the 1804a for this because it is small and nice to use but I suppose an unbalanced cable would be better?  If I'm only using one wire and shield I imagine the quad is not ideal.

The psp2 does have an 1/8 unbalanced output and, gee, it does seem to work. I assumed I would get higher quality output with my own cable on the separate bal outs.

Ok what I would try is this...

If I had a picture of the inside of the preamp I might be able to tell..

Chris

[it-goes-to-eleven]

The main pair of transformers are under that pcb and I doubt there is room for four of that size.

http://taperssection.com/index.php/topic,82497.0.html

[Church-Audio]

The main pair of transformers are under that pcb and I doubt there is room for four of that size.

http://taperssection.com/index.php/topic,82497.0.html

I cant tell much from that picture.. I think they did something strange to the output.. Maybe its not really balanced because if it was.. You would be able to connect pin 3-1 to ground with out issues. I would suspect that they did some strage shit inside that preamp.. I would try cuting the trace between pin 1 and pin 3 and seeing if that does the trick..

Chris
If not I would try pin 3 as hot pin 1 as ground pin 2 N/C